Lithographic apparatus and device manufacturing method utilizing data filtering
Abstract
An apparatus and method are used to form patterns on a substrate. The apparatus comprises a projection system, a patterning device, a low-pass filter, and a data manipulation device. The projection system projects a beam of radiation onto the substrate as an array of sub-beams. The patterning device modulates the sub-beams to substantially produce a requested dose pattern on the substrate. The low-pass filter operates on pattern data derived from the requested dose pattern in order to form a frequency-clipped target dose pattern that comprises only spatial frequency components below a selected threshold frequency. The data manipulation device produces a control signal comprising spot exposure intensities to be produced by the patterning device, based on a direct algebraic least-squares fit of the spot exposure intensities to the frequency-clipped target dose pattern. In various examples, filters can also be used.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1. A lithography apparatus, comprising:
a projection system configured to project a beam of radiation onto a substrate as an array of sub-beams of radiation;
a patterning device configured to modulate the sub-beams of radiation to substantially produce a requested dose pattern on the substrate, the dose pattern being built up from an array of spot exposures in which at least neighboring ones of the spot exposures are imaged incoherently with respect to each other and each of the spot exposures is produced by one of the sub-beams of radiation at a particular time;
a data manipulation device configured to produce a control signal comprising spot exposure intensities to be produced by the patterning device based on a direct algebraic least-squares fit of spot exposure intensities to data derived from the requested dose pattern, wherein the least-squares fit is performed by multiplying a pseudo-inverted form of a point-spread function matrix by a column vector representing the pattern data derived from the requested dose pattern, the point-spread function matrix comprising information about a shape and relative position of the point-spread function of each spot to be exposed on the substrate by one of the sub-beams of radiation at a given time; and
a low-pass filter configured to remove spatial frequency components of a signal above a selected threshold frequency, incorporated offline into the pseudo-inverted form of the point-spread-function matrix, ready for the least-squares fit, by the following operation:
[K] + filtered =F low-pass filter [K] + ,
wherein [K] + and [K] + filtered are the pseudo-inverted form of the point-spread function matrix respectively before and after filtering, and
wherein F low-pass filter is a mathematical definition of the low-pass filter in a spatial domain.
2. The lithography apparatus of claim 1 , further comprising:
a sharpening filter configured to sharpen a target product feature to be formed, defined in a spatial domain by mathematical function F sharp filter , wherein the sharpening filter and the low-pass filter are incorporated offline into the pseudo-inverted form of the point-spread function matrix [K] + to form a filtered point-spread function matrix [K] + filtered via the following operation:
[ K] + filtered =F combined filter [K] + =( F low-pass filter F sharp filter ) [ K] + ,
wherein F combined filter is a mathematical definition of the combined action of the low-pass filter and the sharpening filter in the spatial domain.
3. The lithography apparatus of claim 1 , further comprising:
an image log slope filter configured to control an image log slope of the pattern to be formed on the substrate, defined in the spatial domain by mathematical function F slope filter , wherein the image log slope filter and the low-pass filter are incorporated offline into the point-spread function matrix [K] + to form a filtered point-spread function matrix [K] + filtered via the following operation:
[ K] + filtered =F combined filter [K] + =( F low-pass filter F slope filter ) [ K] + ,
wherein F combined filter is a mathematical definition of the combined action of the low-pass filter and the sharpening filter in the spatial domain.
4. The lithography apparatus of claim 1 , further comprising:
a sharpening filter configured to sharpen a target product feature to be formed, defined in a spatial domain by mathematical function F sharp filter ; and
an image log slope filter configured to control an image log slope of the pattern to be formed on the substrate, defined in the spatial domain by mathematical function F slope filter , wherein the sharpening filter, the image log slope filter, and the low-pass filter are incorporated offline into the point-spread function matrix [K] + to form a filtered point-spread function matrix [K] + filtered via the following operation:
[ K] + filtered =F combined filter [K] + =( F low-pass filter F sharp filter F slope filter ) [ K] + ,
wherein F combined fitter is a mathematical definition of the combined action of the low-pass filter, the sharpening filter, and the image log slope filter in the spatial domain.
5. A lithography apparatus, comprising:
an illumination system configured to condition a beam of radiation;
a projection system configured to project the beam of radiation onto the substrate as an array of sub-beams of radiation;
a patterning device configured to modulate the sub-beams of radiation to substantially produce a requested dose pattern on the substrate, the dose pattern being built up from an array of spot exposures, each of the spot exposure being produced by one of the sub-beams of radiation at a particular time, wherein the radiation intensity of a given sub-beam of radiation is controlled according to an activation state of a corresponding portion of the patterning device;
a dose sensor configured to measure a source intensity of the beam of radiation and a spot intensity of the sub-beams of radiation on the substrate; and
a data manipulation device configured to transform a signal comprising spot exposure radiation doses derived from the requested dose pattern to a control signal representing activation states of the patterning device to produce the requested dose pattern, wherein the transformation corrects for intensity variations measured by the dose sensor as a ratio between the source intensity and the spot intensity and caused by at least one of components of the projection system, components of the illumination system, radiation sources for the illumination system, and components of the patterning device.
6. The lithography apparatus of claim 5 , wherein the data manipulation device comprises:
a memory device configured to store a lookup table, the data manipulation device accessing the lookup table to convert from an activation state for a particular portion of the patterning device to a corresponding control voltage to produce the activation state for that portion, wherein the transformation is performed by changing at least a subset of values in the lookup table.
7. The lithography apparatus of claim 5 , wherein the data manipulation device comprises:
a multiplier configured to convert a spot exposure radiation dose produced by a particular portion of the patterning device to an activation state for that portion, wherein the transformation is performed by changing a gain characteristic of the multiplier.
8. A method, comprising:
projecting a beam of radiation onto a substrate as an array of sub-beams of radiation;
modulating the sub-beams of radiation to substantially produce a requested dose pattern on the substrate, the dose pattern being built up from an array of spot exposures in which at least neighboring ones of the spot exposures are imaged incoherently with respect to each other and each of the spot exposures is produced by one of the sub-beams of radiation at a particular time;
producing a control signal comprising spot exposure intensities to be produced by the modulating based on a direct algebraic least-squares fit of the spot exposure intensities to data derived from the requested dose pattern, wherein the least-squares fit is performed by multiplying a pseudo-inverted form of a point-spread function matrix by a column vector representing the pattern data derived from the requested dose pattern, the point-spread function matrix comprising information about the shape and relative position of the point-spread function of each spot to be exposed on the substrate by one of the sub-beams of radiation at a given time; and
filtering to remove spatial frequency components of a signal above a selected threshold frequency, incorporated offline into the pseudo-inverted form of the point-spread-function matrix, ready for the least-squares fit, by the following operation:
[K] + filtered =F low-pass filter [K] + ,
wherein [K] + and [K] + filtered are the pseudo-inverted form of the point-spread function matrix respectively before and after filtering, and where F low pass filter represents a mathematical definition of the filtering in the spatial domain.
9. A method, comprising:
conditioning a beam of radiation;
detecting a source dose intensity of the beam of radiation;
projecting the beam of radiation onto the substrate as an array of sub-beams of radiation;
detecting a spot dose intensity of the sub-beams of radiation on the substrate;
modulating the sub-beams of radiation to substantially produce a requested dose pattern on the substrate, the dose pattern being built up from an array of spot exposures, each of the spot exposures being produced by one of the sub-beams of radiation at a particular time;
controlling radiation intensity of a given one of the sub-beams of radiation according to an activation state of a corresponding portion of a patterning device that performs the modulating;
transforming a signal comprising spot exposure radiation doses derived from the requested dose pattern to a control signal representing activation states of the patterning device to substantially produce the requested dose pattern; and
modifying the transforming step to correct for intensity variations determined by a ratio of the detected source and spot dose intensities and caused by at least one of components of the projection system, components of the illumination system, radiation sources for the illumination system, and components of the patterning device.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.